Quote:
Originally Posted by darinvg
Doing some comparison between this and SNG Wizard to see the difference. I understand there are spots (especially the bubble) where using ICM isn't a good idea. However why is there a difference in a hand like this:
...
SnG Solver says push 56.3% vs the given call range (24.7%). If I have the BB call 24.7% in SNG Wizard, it says I can push 67.4%.
Just trying to get a base of the difference between the two since all of my previous studying has been with ICM. Thanks!
There are a lot of variables at work in the Predictive Simulation algorithm and the effects (and differences from ICM) can be very non-linear and sometimes hard to intuit.
A few things that stick out in this hand are that 1) there's a shorter stack than us that's about to hit the BB and get even shorter, 2) there is
the short stack that is going to be UTG next hand, and 3) the biggest stack will be in a prime "steal spot" next hand.
All of these things will have an effect on SnG Solver's analysis but wont be accounted for by an ICM based program like SNG Wizard. If you turn the Predictive Simulation to "off" you can see that our push range changes from 56.3% to 61.4% (vs a 24.7% BB call range). That is the difference between PSM and ICM in this spot.
The rest of the difference between 61.4% and what you see in sngwiz might be because of differences in handrankings (Solver's 24.7% is different than Wis'z 24.7%) and because Wiz approximates the contribution from possible "tie pots", whereas Solver calculates tied pots precisely.